Elamur Research⌕
☰
home / papers / 2026 / Why Supervised Fine-Tuning Fails to Learn: A Systematic Study of Incomplete Learning in Large Language Models

Why Supervised Fine-Tuning Fails to Learn: A Systematic Study of Incomplete Learning in Large Language Models

Paper · 2026 · Chao Xue, Yao Wang, Mengqiao Liu et al.

Summary

Supervised fine-tuning (SFT) is usually treated as a reliable way to specialize LLMs. The authors show, however, that converged models still get part of their own training set wrong. They call this the Incomplete Learning Phenomenon (ILP), distinguish it from catastrophic forgetting and machine unlearning, and present what they describe as the first systematic study of it.

Working with Qwen, LLaMA and OLMo2 across several domains and datasets, they build a three-stage, diagnosis-first pipeline:

  1. Fine-tune the model.
  2. Re-run it on the SFT data to flag samples it has not learned.
  3. Calibrate the model or the data.

Each flagged sample is attributed to one of five causes using diagnostic signals, including how consistent its predictions are, how its entropy changes, and how sensitive it is to replay. Cause-specific remedies are then tested as causal interventions. Each yields modest gains, but no single fix resolves every failure, and aggregate metrics can hide stubborn unlearned subsets. This argues for per-sample diagnosis, especially when SFT data is costly to build, as in law or medicine.

Key points

  • The five ILP sources are:
    • gaps in pre-trained knowledge;
    • conflicts between SFT labels and pre-training priors;
    • inconsistencies inside the SFT data;
    • left-side forgetting during sequential fine-tuning;
    • insufficient optimization for rare or compositional patterns.
  • Each cause is paired with a tested remedy:
    • continued pre-training;
    • conflict-aware data processing and scheduling;
    • global shuffling plus dynamic resampling;
    • a Progressive Epoch Increment schedule with validation-based early stopping.
  • On Qwen-7B, the remedies raised accuracy as follows:
    • knowledge-conflict mitigation: 82.3% → 85.1%;
    • left-side-forgetting mitigation on mixed data: 78.5% → 79.8%;
    • adaptive epochs on under-learned tasks: 88.2% → 90.1%.
  • Shuffling plus resampling lifted ROUGE-L on the earliest 10% of the summarization data from 0.41 to 0.53, a gain of about 29%.
  • The authors acknowledge three limitations: reliance on high-quality annotations and external verification tools, risk from noisy supplementary pre-training corpora, and added compute cost.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Peking UniversityBeijing · China315 works · 43,268 citations🏢TencentShenzhen · China163 works · 21,032 citations🎓UNSW SydneySydney · Australia36 works · 1,509 citations🎓University of Electronic Science and Technology of ChinaChengdu · China38 works · 2,088 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 3,894 works

Cite

Cite as Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng and Flora D. Salim. 2026. Why Supervised Fine-Tuning Fails to Learn: A Systematic Study of Incomplete Learning in Large Language Models. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.1393

@inproceedings{xue2026why,
  title = {{Why Supervised Fine-Tuning Fails to Learn: A Systematic Study of Incomplete Learning in Large Language Models}},
  author = {Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng and Flora D. Salim},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1393},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1393}
}

References

147 references, 67 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Rohan Anil, Andrew M Dai, Orhan Firat, Melvin Johnson, Dmitry Lepikhin, Alexandre Passos, Siamak Shakeri, Emanuel Taropa, Paige Bailey, Zhifeng Chen, et al. 2023. Palm 2 technical report. arXiv preprint arXiv:2305.10403.
  2. Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, et al. 2023. Qwen technical report. arXiv preprint arXiv:2309.16609.
  3. Yoshua Bengio, Jérôme Louradour, Ronan Collobert, and Jason Weston. 2009. Curriculum learning. In Proceedings of the 26th annual international conference on machine learning, pages 41–48.
  4. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. 2020. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901.
  5. Yinzhi Cao and Junfeng Yang. 2015. Towards making systems forget with machine unlearning. In 2015 IEEE symposium on security and privacy, pages 463– 480. IEEE. find
  6. Sahil Chaudhary. 2023. Code alpaca: An instruction-following llama model for code generation. https://github.com/sahil280114/codealpaca. ↗
  7. Lusi Chen. 2026a. Beyond external constraints: The missing dimension of ai governance. Available at SSRN 6449738. find
  8. Lusi Chen. 2026b. Testing moral development in ai: An experimental architecture for internal value development in ai governance. Available at SSRN 6472178. find
  9. Yao Chen, Yilong Chen, Yinqi Yang, Junyuan Shang, Zhenyu Zhang, Zefeng Zhang, Shuaiyi Nie, Shuohuan Wang, Yu Sun, Hua Wu, et al. 2026. Sparse growing transformer: Training-time sparse depth allocation via progressive attention looping. arXiv preprint arXiv:2603.23998. find
  10. Yao Chen, Jiawei Sheng, Wenyuan Zhang, and Tingwen Liu. 2025. Improving reasoning capabilities in small models through mixture-of-layers distillation with stepwise attention on key information. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 4952–4971…
  11. Xuxin Cheng, Qianqian Dong, Fengpeng Yue, Tom Ko, Mingxuan Wang, and Yuexian Zou. 2023a. M 3 st: Mix at three levels for speech translation. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE. find
  12. Xuxin Cheng, Zhihong Zhu, Wanshi Xu, Yaowei Li, Hongxiang Li, and Yuexian Zou. 2023b. Accelerating multiple intent detection and slot filling via targeted knowledge distillation. In The 2023 Conference on Empirical Methods in Natural Language Processing. find
  13. Wei-Lin Chiang, Zhuohan Li, Zi Lin, Ying Sheng, Zhanghao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yonghao Zhuang, Joseph E. Gonzalez, Ion Stoica, and Eric P. Xing. 2023. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality. find
  14. Alexandra Chronopoulou, Christos Baziotis, and Alexandros Potamianos. 2019. An embarrassingly simple approach for transfer learning from pretrained language models. arXiv preprint arXiv:1902.10547. find
  15. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Yunxuan Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Alex Castro-Ros, Marie Pellat, Kevin Robinson, Dasha Valter, Sharan… find
  16. Peter Clark, Isaac Cowhey, Oren Etzioni, Tushar Khot, Ashish Sabharwal, Carissa Schoenick, and Oyvind Tafjord. 2018. Think you have solved question answering? try arc, the ai2 reasoning challenge. arXiv:1803.05457v1.
  17. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, et al. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168.
  18. Ganqu Cui, Lifan Lee, Zekun Liu, Cenyuan Yuan, Shijie Li, Yixin Sun, Chi Zhang, YS Tian, Zhaoyang Zhang, Long Li, et al. 2023. Ultrafeedback: Boosting language models with high-quality feedback. arXiv preprint arXiv:2310.01377. find
  19. Chang Dai, Hongyu Shan, Mingyang Song, and Di Liang. 2025. Hope: Hyperbolic rotary positional encoding for stable long-range dependency modeling in large language models. arXiv preprint arXiv:2509.05218. — [arXiv:2509.05218](https://arxiv.org/abs/2509.05218) find
  20. DeepSeek-AI, Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, Damai Dai, Daya Guo, Dejian Yang, Deli Chen, Dongjie Ji, Erhang Li, Fangyun Lin, Fucong Dai, Fuli Luo, Guangbo Hao, Guanting Chen, Guowei Li, H. Zhang, Han Bao, H… find
  21. Guanting Dong, Hongyi Yuan, Keming Lu, Chengpeng Li, Mingfeng Xue, Dayiheng Liu, Wei Wang, Zheng Yuan, Chang Zhou, and Jingren Zhou. 2023. How abilities in large language models are affected by supervised fine-tuning data composition. arXiv preprint arXiv:2310.05492.
  22. Edoardo Federici. 2022. sentence-bert-base, sentence-transformer for italian. find
  23. Zichu Fei, Qi Zhang, Tao Gui, Di Liang, Sirui Wang, Wei Wu, and Xuan-Jing Huang. 2022. Cqg: A simple and effective controlled generation framework for multi-hop question generation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pa… find
  24. Dayuan Fu, Jianzhao Huang, Siyuan Lu, Guanting Dong, Yejie Wang, Keqing He, and Weiran Xu. 2024. Preact: Predicting future in react enhances agent’s planning ability. Preprint, arXiv:2402.11534. — [arXiv:2402.11534](https://arxiv.org/abs/2402.11534) find
  25. Leo Gao, John Schulman, and Jacob Hilton. 2022. Scaling laws for reward model overoptimization. Preprint, arXiv:2210.10760.
  26. Ziyuan Gao, Di Liang, Xianjie Wu, Philippe Morel, and Minlong Peng. 2026. Decorl: Decoupling reasoning chains via parallel sub-step generation and cascaded reinforcement for interpretable and scalable rlhf. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 40, pages 30789–3079… find
  27. Aaron Gokaslan, Vanya Cohen, Ellie Pavlick, and Stefanie Tellex. 2019. Openwebtext corpus. http://Skylion007.github.io/OpenWebTextCorpus. ↗
  28. Zhibin Gou, Zhihong Shao, Yeyun Gong, Yelong Shen, Yujiu Yang, Minlie Huang, Nan Duan, and Weizhu Chen. 2024. Tora: A tool-integrated reasoning agent for mathematical problem solving. Preprint, arXiv:2309.17452.
  29. Neel Guha, Julian Nyarko, Daniel E. Ho, Christopher Ré, Adam Chilton, Aditya Narayana, Alex Chohlas-Wood, Austin Peters, Brandon Waldon, Daniel N. Rockmore, Diego Zambrano, Dmitry Talisman, Enam Hoque, Faiz Surani, Frank Fagan, Galit Sarfaty, Gregory M. Dickinson, Haggai Porat, Jason Hegland, Jessic… — LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models
  30. Tao Gui, Qi Zhang, Jingjing Gong, Minlong Peng, Di Liang, Keyu Ding, and Xuan-Jing Huang. 2018. Transferring from formal newswire domain with hypernet for twitter pos tagging. In Proceedings of the 2018 conference on empirical methods in natural language processing, pages 2540–2549. find
  31. Weiyang Guo, Zesheng Shi, Liye Zhao, Jiayuan Ma, Zeen Zhu, Junxian He, Min Zhang, and Jing Li. 2026a. E3-tir: Enhanced experience exploitation for tool-integrated reasoning. Preprint, arXiv:2604.09455. — [arXiv:2604.09455](https://arxiv.org/abs/2604.09455) find
  32. Weiyang Guo, Zesheng Shi, Zeen Zhu, Yuan Zhou, Min Zhang, and Jing Li. 2026b. Backdoors in rlvr: Jailbreak backdoors in llms from verifiable reward. Preprint, arXiv:2604.09748.
  33. Suchin Gururangan, Ana Marasović, Swabha Swayamdipta, Kyle Lo, Iz Beltagy, Doug Downey, and Noah A. Smith. 2020. Don’t stop pretraining: Adapt language models to domains and tasks. Preprint, arXiv:2004.10964. find
  34. Dan Hendrycks, Collin Burns, Steven Basart, Andrew Critch, Jerry Li, Dawn Song, and Jacob Steinhardt. 2021a. Aligning ai with shared human values. Proceedings of the International Conference on Learning Representations (ICLR).
  35. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021b. Measuring massive multitask language understanding. Proceedings of the International Conference on Learning Representations (ICLR).
  36. Geoffrey Hinton. 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531.
  37. Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, and Laurent Sifre. 2022. Training compute-optimal large language models. Preprint, arXiv:2203.15556.
  38. Xia Hou, Qifeng Li, Jian Yang, Tongliang Li, Linzheng Chai, Xianjie Wu, Hangyuan Ji, Zhoujun Li, Jixuan Nie, Jingbo Dun, et al. 2024a. Raw text is all you need: Knowledge-intensive multi-turn instruction tuning for large language model. arXiv preprint arXiv:2407.03040. — [arXiv:2407.03040](https://arxiv.org/abs/2407.03040) find
  39. Yupeng Hou, Junjie Zhang, Zihan Lin, Hongyu Lu, Ruobing Xie, Julian McAuley, and Wayne Xin Zhao. 2024b. Large language models are zero-shot rankers for recommender systems. In European Conference on Information Retrieval, pages 364–381. Springer.
  40. Jianpeng Hu, Chao Xue, Chunqing Yu, Jiacheng Xu, and Chengxiang Tan. 2025. Joint learning event-specific probe and argument library with differential optimization for document-level multi-event extraction. In Findings of the Association for Computational Linguistics: NAACL 2025, pages 714–726. find
  41. Liang Hu and Yinru Shen. 2026. A predictive analytics approach for forecasting global stock index returns using deep learning techniques. Decision Analytics Journal, page 100685. find
  42. Peichun Hua, Hao Li, Shanghao Shi, Zhiyuan Yu, and Ning Zhang. 2025. Rethinking jailbreak detection of large vision language models with representational contrastive scoring. Preprint, arXiv:2512.12069.
  43. Fanding Huang, Guanbo Huang, Xiao Fan, Yi He, Xiao Liang, Xiao Chen, Qinting Jiang, Faisal Nadeem Khan, Jingyan Jiang, and Zhi Wang. 2026. Semantic-space exploration and exploitation in rlvr for llm reasoning. Preprint, arXiv:2509.23808. find
  44. Pranab Islam, Anand Kannappan, Douwe Kiela, Rebecca Qian, Nino Scherrer, and Bertie Vidgen. 2023. Financebench: A new benchmark for financial question answering. Preprint, arXiv:2311.11944. — [arXiv:2311.11944](https://arxiv.org/abs/2311.11944) find
  45. Hongru Ji, Yuyin Fan, Meng Zhao, Xianghua Li, Lianwei Wu, and Chao Gao. 2026. Stride-ed: A strategy-grounded stepwise reasoning framework for empathetic dialogue systems. Preprint, arXiv:2604.07100.
  46. Yuxuan Jiang and Francis Ferraro. 2026. Beyond math: Stories as a testbed for memorization-constrained reasoning in llms. In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5590–5607.
  47. Yuxuan Jiang, Dawei Li, and Frank Ferraro. 2025. Drp: Distilled reasoning pruning with skill-aware step decomposition for efficient large reasoning models. arXiv preprint arXiv:2505.13975.
  48. Di Jin, Eileen Pan, Nassim Oufattole, Wei-Hung Weng, Hanyi Fang, and Peter Szolovits. 2020. What disease does this patient have? a large-scale open domain question answering dataset from medical exams. arXiv preprint arXiv:2009.13081.
  49. Mandar Joshi, Eunsol Choi, Daniel Weld, and Luke Zettlemoyer. 2017. triviaqa: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension. arXiv e-prints, arXiv:1705.03551.
  50. Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. 2020. Scaling laws for neural language models. arXiv preprint arXiv:2001.08361.
  51. Daniel Khashabi, Sewon Min, Asqar Saparov, Hannaneh Hajishirzi, Wen-tau Yih, and Peter Clark. 2020. Unifiedqa: Crossing format boundaries with a single qa system. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1896– 1907.
  52. Yuta Koreeda and Christopher D Manning. 2021. Contractnli: A dataset for document-level natural language inference for contracts. arXiv preprint arXiv:2110.01799.
  53. Suhas Kotha, Jacob Mitchell Springer, and Aditi Raghunathan. 2023. Understanding catastrophic forgetting in language models via implicit inference. arXiv preprint arXiv:2309.10105.
  54. Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Jacob Devlin, Kenton Lee, Kristina Toutanova, Llion Jones, Matthew Kelcey, Ming-Wei Chang, Andrew M. Dai, Jakob Uszkoreit, Quoc Le, and Slav Petrov. 2019a. Natural…
  55. Tom Kwiatkowski, Jennimaria Palomaki, Olivia Redfield, Michael Collins, Ankur Parikh, Chris Alberti, Danielle Epstein, Illia Polosukhin, Matthew Kelcey, Jacob Devlin, Kenton Lee, Kristina N. Toutanova, Llion Jones, Ming-Wei Chang, Andrew Dai, Jakob Uszkoreit, Quoc Le, and Slav Petrov. 2019b. Natural…
  56. Shanglin Lei, Guanting Dong, Xiaoping Wang, Keheng Wang, and Sirui Wang. 2023. Instructerc: Reforming emotion recognition in conversation with a retrieval multi-task llms framework. Preprint, arXiv:2309.11911. — [arXiv:2309.11911](https://arxiv.org/abs/2309.11911) find
  57. Brian Lester, Rami Al-Rfou, and Noah Constant. 2021. The power of scale for parameter-efficient prompt tuning. arXiv preprint arXiv:2104.08691.
  58. Bo Li, Di Liang, and Zixin Zhang. 2024a. Co-mateformer: Combined attention transformer for semantic sentence matching. arXiv preprint arXiv:2412.07220. — [arXiv:2412.07220](https://arxiv.org/abs/2412.07220) find
  59. Chen-An Li and Hung-Yi Lee. 2024. Examining forgetting in continual pre-training of aligned large language models. arXiv preprint arXiv:2401.03129. — [arXiv:2401.03129](https://arxiv.org/abs/2401.03129) find
  60. Junchen Li, Chao Qi, Rongzheng Wang, Qizhi Chen, Liang Xu, Di Liang, Bob Simons, and Shuang Liang. 2026. When safety becomes a vulnerability: Exploiting llm alignment homogeneity for transferable blocking in rag. arXiv preprint arXiv:2603.03919. — [arXiv:2603.03919](https://arxiv.org/abs/2603.03919) find
  61. Liang Li, Qisheng Liao, Meiting Lai, Di Liang, and Shangsong Liang. 2024b. Local and global: Text matching via syntax graph calibration. In ICASSP 2024-2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 11571–11575. IEEE. find
  62. Ming Li, Yong Zhang, Shwai He, Zhitao Li, Hongyu Zhao, Jianzong Wang, Ning Cheng, and Tianyi Zhou. 2024c. Superfiltering: Weak-to-strong data filtering for fast instruction-tuning. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pag…
  63. Di Liang, Fubao Zhang, Qi Zhang, and Xuan-Jing Huang. 2019a. Asynchronous deep interaction network for natural language inference. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNL… find
  64. Di Liang, Fubao Zhang, Weidong Zhang, Qi Zhang, Jinlan Fu, Minlong Peng, Tao Gui, and Xuanjing Huang. 2019b. Adaptive multi-attention network incorporating answer information for duplicate question detection. In Proceedings of the 42nd international ACM SIGIR conference on research and development i… find
  65. Xiaobo Liang, Lijun Wu, Juntao Li, Yue Wang, Qi Meng, Tao Qin, Wei Chen, Min Zhang, and Tie-Yan Liu. 2021. R-drop: Regularized dropout for neural networks. Preprint, arXiv:2106.14448. — [arXiv:2106.14448](https://arxiv.org/abs/2106.14448) find
  66. Bo Liu, Liming Zhan, Zexin Lu, Yujie Feng, Lei Xue, and Xiao-Ming Wu. 2024a. How good are llms at out-of-distribution detection? Preprint, arXiv:2308.10261. find
  67. Peiyang Liu, Ziqiang Cui, Di Liang, and Wei Ye. 2025a. Who stole your data? a method for detecting unauthorized rag theft. arXiv preprint arXiv:2510.07728. — [arXiv:2510.07728](https://arxiv.org/abs/2510.07728) find
  68. Wei Liu, Weihao Zeng, Keqing He, Yong Jiang, and Junxian He. 2024b. What makes good data for alignment? a comprehensive study of automatic data selection in instruction tuning. Preprint, arXiv:2312.15685.
  69. Xiaoyu Liu, Xiaoyu Guan, Di Liang, and Xianjie Wu. 2026. Dpi: Exploiting parameter heterogeneity for interference-free fine-tuning. arXiv preprint arXiv:2601.17777. — [arXiv:2601.17777](https://arxiv.org/abs/2601.17777) find
  70. Xiaoyu Liu, Di Liang, Hongyu Shan, Peiyang Liu, Yonghao Liu, Muling Wu, Yuntao Li, Xianjie Wu, Li Miao, Jiangrong Shen, et al. 2025b. Structural reward model: Enhancing interpretability, efficiency, and scalability in reward modeling. In Proceedings of the 2025 Conference on Empirical Methods in Nat… find
  71. Yonghao Liu, Mengyu Li, Di Liang, Ximing Li, Fausto Giunchiglia, Lan Huang, Xiaoyue Feng, and Renchu Guan. 2024c. Resolving word vagueness with scenario-guided adapter for natural language inference. arXiv preprint arXiv:2405.12434. — [arXiv:2405.12434](https://arxiv.org/abs/2405.12434) find
  72. Yonghao Liu, Di Liang, Fang Fang, Sirui Wang, Wei Wu, and Rui Jiang. 2023a. Time-aware multiway adaptive fusion network for temporal knowledge graph question answering. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE. find
  73. Yonghao Liu, Di Liang, Mengyu Li, Fausto Giunchiglia, Ximing Li, Sirui Wang, Wei Wu, Lan Huang, Xiaoyue Feng, and Renchu Guan. 2023b. Local and global: Temporal question answering via information fusion. In IJCAI, pages 5141–5149. find
  74. Keming Lu, Hongyi Yuan, Zheng Yuan, Runji Lin, Junyang Lin, Chuanqi Tan, and Chang Zhou. 2023. \# instag: Instruction tagging for diversity and complexity analysis. arXiv preprint arXiv:2308.07074.
  75. Ziyang Luo, Can Xu, Pu Zhao, Qingfeng Sun, Xiubo Geng, Wenxiang Hu, Chongyang Tao, Jing Ma, Qingwei Lin, and Daxin Jiang. 2023. Wizardcoder: Empowering code large language models with evol-instruct. arXiv preprint arXiv:2306.08568. find
  76. Ruotian Ma, Yiding Tan, Xin Zhou, Xuanting Chen, Di Liang, Sirui Wang, Wei Wu, Tao Gui, and Qi Zhang. 2022. Searching for optimal subword tokenization in cross-domain ner. arXiv preprint arXiv:2206.03352. — [arXiv:2206.03352](https://arxiv.org/abs/2206.03352) find
  77. Andrew L. Maas, Raymond E. Daly, Peter T. Pham, Dan Huang, Andrew Y. Ng, and Christopher Potts. 2011. Learning word vectors for sentiment analysis. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 142–150, Portland, Oregon…
  78. Mark Mazumder, Colby Banbury, Xiaozhe Yao, Bojan Karlaš, William Gaviria Rojas, Sudnya Diamos, Greg Diamos, Lynn He, Alicia Parrish, Hannah Rose Kirk, et al. 2023. Dataperf: Benchmarks for data-centric ai development. Advances in Neural Information Processing Systems, 36:5320–5347. find
  79. Michael McCloskey and Neal J Cohen. 1989. Catastrophic interference in connectionist networks: The sequential learning problem. In Psychology of learning and motivation, volume 24, pages 109–165. Elsevier. find
  80. Subhabrata Mukherjee, Arindam Mitra, Ganesh Jawahar, Sahaj Agarwal, Hamid Palangi, and Ahmed Awadallah. 2023. Orca: Progressive learning from complex explanation traces of gpt-4. Preprint, arXiv:2306.02707.
  81. Jishnu Mukhoti, Sanyam Rajeswar, Harsh Singh, Koushik Rajan, Sebastian Ruder, Pratik Kumar, Aditya Raghunathan, Anoop Kunchukuttan, Digvijay Kumar, and Sunita Sarawagi. 2023. Fine-tuning can cripple your foundation model; preserving features may be the solution. arXiv preprint arXiv:2308.13320. — [arXiv:2308.13320](https://arxiv.org/abs/2308.13320) find
  82. Shashi Narayan, Shay B. Cohen, and Mirella Lapata. 2018. Don’t give me the details, just the summary! topic-aware convolutional neural networks for extreme summarization. ArXiv, abs/1808.08745.
  83. Team OLMo, Pete Walsh, Luca Soldaini, Dirk Groeneveld, Kyle Lo, Shane Arora, Akshita Bhagia, Yuling Gu, Shengyi Huang, Matt Jordan, Nathan Lambert, Dustin Schwenk, Oyvind Tafjord, Taira Anderson, David Atkinson, Faeze Brahman, Christopher Clark, Pradeep Dasigi, Nouha Dziri, Allyson Ettinger, Michal… find
  84. OpenAI, Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, Red Avila, Igor Babuschkin, Suchir Balaji, Valerie Balcom, Paul Baltescu, Haiming Bao, Mohammad Bavarian, Jeff Belgum, Irwan Bello, Ja… — GPT-4 Technical Report
  85. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. 2022. Train…
  86. Ankit Pal, Logesh Kumar Umapathi, and Malaikannan Sankarasubbu. 2022. Medmcqa: A large-scale multi-subject multi-choice dataset for medical domain question answering. In Proceedings of the Conference on Health, Inference, and Learning, volume 174 of Proceedings of Machine Learning Research, pages 24…
  87. German I Parisi, Ronald Kemker, Jose L Part, Christopher Kanan, and Stefan Wermter. 2019. Continual lifelong learning with neural networks: A review. Neural networks, 113:54–71.
  88. Emmanouil Antonios Platanios, Otilia Stretcu, Graham Neubig, Barnabas Poczos, and Tom M Mitchell. 2019. Competence-based curriculum learning for neural machine translation. arXiv preprint arXiv:1903.09848. find
  89. Edoardo Maria Ponti, Clara Vania, Goran Glavaš, Olga Majewska, Zining Wu, Jannis Lin, Ivan Vulic, and Anna Korhonen. 2023. Fine-tuning language models for specific tasks can be harmful. arXiv preprint arXiv:2310.09419. — [arXiv:2310.09419](https://arxiv.org/abs/2310.09419) find
  90. David Rein, Betty Li Hou, Asa Cooper Stickland, Jackson Petty, Richard Yuanzhe Pang, Julien Dirani, Julian Michael, and Samuel R. Bowman. 2024. GPQA: A graduate-level google-proof q&a benchmark. In First Conference on Language Modeling.
  91. Sebastian Ruder. 2017. An overview of multi-task learning in deep neural networks. arXiv preprint arXiv:1706.05098.
  92. Victor Sanh, Lysandre Debut, Julien Chaumond, and Thomas Wolf. 2019. Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108.
  93. Victor Sanh, Albert Webson, Thomas Wolf, and Alexander M. Rush. 2022. Multitask prompted training enables zero-shot task generalization. Preprint, arXiv:2110.08207.
  94. Abigail See, Peter J. Liu, and Christopher D. Manning. 2017. Get to the point: Summarization with pointer-generator networks. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1073– 1083, Vancouver, Canada. Association for Compu… find
  95. Burr Settles. 2009. Active learning literature survey.
  96. Connor Shorten and Taghi M Khoshgoftaar. 2019. A survey on image data augmentation for deep learning. Journal of big data, 6(1):1–48. find
  97. Karan Singhal, Shekoofeh Azizi, Tao Tu, S. Sara Mahdavi, Jason Wei, Hyung Won Chung, Nathan Scales, Yun Liu, Alvin Rajkomar, Joelle Barral, Christopher Semturs, Alan Karthikesalingam, and Vivek Natarajan. 2022. Large language models encode clinical knowledge. Preprint, arXiv:2212.13138.
  98. Lutz Prechelt. 2002. Early stopping-but when? In Neural Networks: Tricks of the trade, pages 55–69. Springer. find
  99. Jian Song, Di Liang, Rumei Li, Yuntao Li, Sirui Wang, Minlong Peng, Wei Wu, and Yongxin Yu. 2022. Improving semantic matching through dependency-enhanced pre-trained model with adaptive fusion. arXiv preprint arXiv:2210.08471. find
  100. Qi Qian, Muling Wu, Zisu Huang, Wenhao Liu, Changze Lv, Xiaohua Wang, Zhenghua Wang, Zhengkang Guo, Zhibo Xu, Lina Chen, et al. Adaptive curriculum strategies: Stabilizing reinforcement learning for large language models. find
  101. Xiaoshuai Song, Keqing He, Pei Wang, Guanting Dong, Yutao Mou, Jingang Wang, Yunsen Xian, Xunliang Cai, and Weiran Xu. 2023. Large language models meet open-world intent discovery and recognition: An evaluation of chatgpt. Preprint, arXiv:2310.10176. find
  102. Pranav Rajpurkar, Jian Zhang, Konstantin Lopyrev, and Percy Liang. 2016. SQuAD: 100,000+ questions for machine comprehension of text. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 2383–2392, Austin, Texas. Association for Computational Linguistics.
  103. Siva Reddy, Danqi Chen, and Christopher D. Manning. 2019. CoQA: A conversational question answering challenge. Transactions of the Association for Computational Linguistics, 7:249–266.
  104. Xiaoshuai Song, Zhengyang Wang, Keqing He, Guanting Dong, Yutao Mou, Jinxu Zhao, and Weiran Xu. 2024. Knowledge editing on black-box large language models. Preprint, arXiv:2402.08631. — [arXiv:2402.08631](https://arxiv.org/abs/2402.08631) find
  105. Mirac Suzgun, Nathan Scales, Nathanael Schärli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc V Le, Ed H Chi, Denny Zhou, , and Jason Wei. 2022. Challenging big-bench tasks and whether chain-of-thought can solve them. arXiv preprint arXiv:2210.09261.
  106. Alon Talmor, Jonathan Herzig, Nicholas Lourie, and Jonathan Berant. 2019. CommonsenseQA: A question answering challenge targeting commonsense knowledge. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,…
  107. Rohan Taori, Ishaan Gulrajani, Tianyi Zhang, Yann Dubois, Xuechen Li, Carlos Guestrin, Percy Liang, and Tatsunori B. Hashimoto. 2023. Stanford alpaca: An instruction-following llama model. https://github.com/tatsu-lab/stanford_alpaca. ↗
  108. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller,… — Llama 2: Open Foundation and Fine-Tuned Chat Models
  109. Guan Wang, Sijie Cheng, Xianyuan Zhan, Xiangang Li, Sen Song, and Yang Liu. 2023a. Openchat: Advancing open-source language models with mixed-quality data. Preprint, arXiv:2309.11235.
  110. Pei Wang, Yejie Wang, Muxi Diao, Keqing He, Guanting Dong, and Weiran Xu. 2024. Multi-perspective consistency enhances confidence estimation in large language models. Preprint, arXiv:2402.11279. — [arXiv:2402.11279](https://arxiv.org/abs/2402.11279) find
  111. Rongzheng Wang, Yihong Huang, Muquan Li, Jiakai Li, Di Liang, Bob Simons, Pei Ke, Shuang Liang, and Ke Qin. 2026a. Rethinking llm-driven heuristic design: Generating efficient and specialized solvers via dynamics-aware optimization. arXiv preprint arXiv:2601.20868. — [arXiv:2601.20868](https://arxiv.org/abs/2601.20868) find
  112. Sirui Wang, Di Liang, Jian Song, Yuntao Li, and Wei Wu. 2022. Dabert: Dual attention enhanced bert for semantic matching. arXiv preprint arXiv:2210.03454. find
  113. Yao Wang, Di Liang, and Minlong Peng. 2025. Not all parameters are created equal: Smart isolation boosts fine-tuning performance. arXiv preprint arXiv:2508.21741. find
  114. Yizhong Wang, Hamish Ivison, Pradeep Dasigi, Jack Hessel, Tushar Khot, Khyathi Raghavi Chandu, David Wadden, Kelsey MacMillan, Noah A Smith, Iz Beltagy, et al. 2023b. How far can camels go? exploring the state of instruction tuning on open resources. arXiv preprint arXiv:2306.04751.
  115. Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, and Hannaneh Hajishirzi. 2023c. Self-instruct: Aligning language models with self-generated instructions. Preprint, arXiv:2212.10560. find
  116. Yu Wang, Emmanuele Chersoni, and Chu-Ren Huang. 2026b. This one or that one? a study on accessibility via demonstratives with multimodal large language models. In Language Resources and Evaluation Conference 2026. European Language Resources Association (ELRA).
  117. Yuxiang Wei, Zhe Wang, Jiawei Liu, Yifeng Ding, and Lingming Zhang. 2023. Magicoder: Source code is all you need. Preprint, arXiv:2312.02120. find
  118. Wuzhenghong Wen, Chao Xue, Su Pan, Yuwei Sun, and Minlong Peng. 2026. Reinforcement learning enhanced multi-hop reasoning for temporal knowledge question answering. arXiv preprint arXiv:2601.01195. — [arXiv:2601.01195](https://arxiv.org/abs/2601.01195) find
  119. Adina Williams, Nikita Nangia, and Samuel Bowman. 2018. A broad-coverage challenge corpus for sentence understanding through inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Pa…
  120. Muling Wu, Qi Qian, Wenhao Liu, Xiaohua Wang, Zisu Huang, Di Liang, LI Miao, Shihan Dou, Changze Lv, Zhenghua Wang, et al. 2025a. Progressive mastery: Customized curriculum learning with guided prompting for mathematical reasoning. arXiv preprint arXiv:2506.04065. — [arXiv:2506.04065](https://arxiv.org/abs/2506.04065) find
  121. Shaohua Wu, Xudong Zhao, Tong Yu, Rongguo Zhang, Chong Shen, Hongli Liu, Feng Li, Hong Zhu, Jiangang Luo, Liang Xu, et al. 2021. Yuan 1.0: Large-scale pre-trained language model in zero-shot and few-shot learning. arXiv preprint arXiv:2110.04725. find
  122. Xianjie Wu, Di Liang, Jian Yang, Xianfu Cheng, LinZheng Chai, Tongliang Li, Liqun Yang, and Zhoujun Li. 2025b. Breaking size barrier: Enhancing reasoning for large-size table question answering. In International Conference on Database Systems for Advanced Applications, pages 241–256. Springer. find
  123. Xianjie Wu, Jian Yang, Linzheng Chai, Ge Zhang, Jiaheng Liu, Xeron Du, Di Liang, Daixin Shu, Xianfu Cheng, Tianzhen Sun, et al. 2025c. Tablebench: A comprehensive and complex benchmark for table question answering. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 39, pages 25… find
  124. Xianjie Wu, Jian Yang, Tongliang Li, Shiwei Zhang, Yiyang Du, LinZheng Chai, Di Liang, and Zhoujun Li. 2025d. Unleashing potential of evidence in knowledge-intensive dialogue generation. In ICASSP 2025-2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5.… find
  125. Can Xu, Qingfeng Sun, Kai Zheng, Xiubo Geng, Pu Zhao, Jiazhan Feng, Chongyang Tao, and Daxin Jiang. 2023. Wizardlm: Empowering large language models to follow complex instructions. Preprint, arXiv:2304.12244.
  126. Scaling relationship on learning mathematical reasoning with large language models. Preprint, arXiv:2308.01825.
  127. Zheng Yuan, Hongyi Yuan, Chuanqi Tan, Wei Wang, Songfang Huang, and Fei Huang. 2023b. Rrhf: Rank responses to align language models with human feedback without tears. Preprint, arXiv:2304.05302.
  128. Xiang Yue, Xingwei Qu, Ge Zhang, Yao Fu, Wenhao Huang, Huan Sun, Yu Su, and Wenhu Chen. 2023. Mammoth: Building math generalist models through hybrid instruction tuning. arXiv preprint arXiv:2309.05653.
  129. Chao Xue and Ziyuan Gao. 2025. Structcoh: Structured contrastive learning for context-aware text semantic matching. In Pacific Rim International Conference on Artificial Intelligence, pages 300–315. Springer. find
  130. Xiang Yue, Tuney Zheng, Ge Zhang, and Wenhu Chen. 2024. Mammoth2: Scaling instructions from the web. Preprint, arXiv:2405.03548.
  131. Chao Xue, Di Liang, Pengfei Wang, and Jing Zhang. 2024. Question calibration and multi-hop modeling for temporal question answering. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pages 19332–19340. find
  132. Hengyuan Zhang, Shiping Yang, Xiao Liang, Chenming Shang, Yuxuan Jiang, Chaofan Tao, Jing Xiong, Hayden Kwok-Hay So, Ruobing Xie, Angel X Chang, et al. 2025. Find your optimal teacher: Personalized data synthesis via router-guided multi-teacher distillation. arXiv preprint arXiv:2510.10925.
  133. Chao Xue, Di Liang, Sirui Wang, Jing Zhang, and Wei Wu. 2023a. Dual path modeling for semantic matching by perceiving subtle conflicts. In ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE. find
  134. Shengyu Zhang, Linfeng Dong, Xiaoya Li, Sen Zhang, Xiaofei Sun, Shuhe Wang, Jiwei Li, Runyi Hu, Tianwei Zhang, Fei Wu, and Guoyin Wang. 2024. Instruction tuning for large language models: A survey. Preprint, arXiv:2308.10792.
  135. Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng, and Flora D. Salim. 2026. Reason only when needed: Efficient generative reward modeling via model-internal uncertainty. Preprint, arXiv:2604.10072.
  136. Mingfeng Xue, Dayiheng Liu, Kexin Yang, Guanting Dong, Wenqiang Lei, Zheng Yuan, Chang Zhou, and Jingren Zhou. 2023b. Occuquest: Mitigating occupational bias for inclusive large language models. Preprint, arXiv:2310.16517. — [arXiv:2310.16517](https://arxiv.org/abs/2310.16517) find
  137. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, and Karthik Narasimhan. 2023a. Tree of thoughts: Deliberate problem solving with large language models. Preprint, arXiv:2305.10601. find
  138. Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. 2023b. React: Synergizing reasoning and acting in language models. Preprint, arXiv:2210.03629.
  139. Lifan Yuan, Yangyi Chen, Ganqu Cui, Hongcheng Gao, Fangyuan Zou, Xingyi Cheng, Heng Ji, Zhiyuan Liu, and Maosong Sun. 2024. Revisiting out-of-distribution robustness in nlp: Benchmarks, analysis, and llms evaluations. Advances in Neural Information Processing Systems, 36. find
  140. Zheng Yuan, Hongyi Yuan, Chengpeng Li, Guanting Dong, Chuanqi Tan, and Chang Zhou. 2023a. find
  141. Xiang Zhang, Junbo Jake Zhao, and Yann LeCun. 2015. Character-level convolutional networks for text classification. In NIPS.
  142. Xiao Zhang and Ji Wu. 2024. Dissecting learning and forgetting in language model finetuning. In The Twelfth International Conference on Learning Representations. find
  143. Yang Zhao, Li Du, Xiao Ding, Kai Xiong, Ting Liu, and Bing Qin. 2024. Supervised fine-tuning achieve rapid task adaption via alternating attention head activation patterns. arXiv preprint arXiv:2409.15820. — [arXiv:2409.15820](https://arxiv.org/abs/2409.15820) find
  144. Rui Zheng, Rong Bao, Yuhao Zhou, Di Liang, Sirui Wang, Wei Wu, Tao Gui, Qi Zhang, and Xuanjing Huang. 2022. Robust lottery tickets for pre-trained language models. arXiv preprint arXiv:2211.03013. find
  145. Wanjun Zhong, Ruixiang Cui, Yiduo Guo, Yaobo Liang, Shuai Lu, Yanlin Wang, Amin Saied, Weizhu Chen, and Nan Duan. 2023. Agieval: A human-centric benchmark for evaluating foundation models. Preprint, arXiv:2304.06364.
  146. Chunting Zhou, Pengfei Liu, Puxin Xu, Srini Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, et al. 2023a. Lima: Less is more for alignment. arXiv preprint arXiv:2305.11206. find
  147. Jeffrey Zhou, Tianjian Lu, Swaroop Mishra, Siddhartha Brahma, Sujoy Basu, Yi Luan, Denny Zhou, and Le Hou. 2023b. Instruction-following evaluation for large language models. Preprint, arXiv:2311.07911.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/xue-2026-why-supervised-fine-tuning-fails/